Assessment of gear pit monitoring system based on order analysis

Abstract

Conventional monitoring methods are based on stationary assumptions, thus they are not applicable to the diagnosis of gear working under varying speed and/or non-stationary conditions. In order to extend the conventional diagnostic methods to speed variation cases of the industrial application expressly for maintenance decisions, order analyses techniques are proposed. Order analyses methods are introduced to extract the harmonic orders of pitted gears from the response signals under constant and variable speed conditions, and the FFT-based analyses are performed in order spectra. Order cepstrum is presented for gear fault diagnosis under steady state operation condition to measure the ability of this type of order analysis as a monitoring technique at different operation conditions. The effectiveness of the proposed methods is demonstrated by real vibration signals collected from the vehicle gearbox. Analysed results show that the proposed methods could identify faulty gears effectively and accurately under different operation conditions.